The following problems will be investigated: (1) Does the preferential expression of anti-arsonate antibodies in the allotype al of ala3 heterozygous rabbits depend on the protein carrier used in our experiments? We will replace bovine serum albumin by sheep serum gamma-globulin as protein carrier. (2) Is the same preferential expression found when azobenzenearsonate-protein is injected into a2a3 heterozygotes? (3) Do the haptens used in our experiments interact with normal rabbit serum immunoglobulins? The answers to these questions may give us more insight into the mechanism of the antigenic stimulus.